Downloaded from genesdev.cshlp.org on September 27, 2021 - Published by Cold Spring Harbor Laboratory Press

Induction of specific macronuclear developmental mutations by microinjection of a cloned telomeric gene in Paramecium primaurelia

Eric Meyer Laboratoire de Genetique Moleculaire, Ecole Normale Superieure, 75230 Paris Cedex, France

In Paramecium, the differentiation of a highly polyploid macronucleus from a diploid nucleus is accompanied by an extensive reorganization of the genome, involving reduction in chromosome size and formation of new telomeres at heterogeneous, but reproducible, positions. The results presented here, as well as work by others, indicate that telomere addition regions are not strictly determined by the micronuclear sequence, but are at least partially controlled by the old macronucleus. It is shown that microinjecting a high copy number of a plasmid containing the G surface antigen gene into the macronucleus of wild-type cells specifically modifies the processing of the G gene-bearing micronuclear chromosome at the following autogamy. Telomeric repeats are added upstream of the gene, rather than at their wild-type position 5 kb downstream of its 3' end, resulting in the deletion of the gene from the new macronucleus. This macronuclear mutation is unstable at the following autogamy, giving rise to many different telomere addition regions in different postautogamous clones. However, after several successive autogamies, cell lines can be obtained in which the telomeres reproducibly form in the same region. In crosses with wild-type cells, these macronuclear mutations show cytoplasmic inheritance; the micronuclei of the mutants are shown to be fully functional. The implications for the mechanism of choice of telomere addition sites are discussed. [Key Words: Developmental mutants; genome reorganization; macronuclear development; Paramecium primaurelia; telomere formation] Received November 25, 1991; accepted December 23, 1991.

Paramecium, like other ciliates, exhibits an interesting division to the two daughter cells. Thus, autogamy pro­ phenomenon of nuclear differentiation. Each cell con­ duces two cells with independently developed macronu­ tains two kinds of nuclei with markedly different struc­ clei, which are called caryonides; conjugation produces tures and functions: The micronuclei are small, diploid four different caryonides, two from each exconjugant cell nuclei that are apparently not transcribed but only serve (for a complete description, see Sonneborn 1974). to transmit genetic information from one sexual gener­ The differentiation of a new somatic macronucleus ation to the next through the processes of conjugation or from a diploid germinal nucleus is accompanied by an autogamy; the macronucleus is a large, highly polyploid extensive and reproducible reorganization of the ge­ (~800n) nucleus where most, if not all, of the cell's tran­ nome, involving amplification to the final ploidy level, scription takes place. During sexual events, the micro- chromosome fragmentation, and elimination of specific nuclei undergo meiotic divisions to produce two identi­ sequences. These processes occur to various extents in cal haploid pronuclei, while the macronucleus breaks different species of ciliates (for review, see Yao 1989). In down into fragments that are diluted with successive Paramecium primaurelia, reduction of chromosome size fissions, and presumably degraded. In the case of conju­ obviously occurs during macronuclear development, be­ gation, the two mating cells exchange one of their hap­ cause the size of macronuclear chromosomes ranges loid pronuclei; fusion of the nuclei then gives two ex- from 50 to 800 kb (Caron and Meyer 1989), with an av­ conjugant cells with identical diploid zygotic nuclei. If erage of about 300 kb (Freer and Freer 1979), whereas the only one mating type is present, the cell undergoes au­ 30 to 60 micronuclear chromosome pairs (Sonneborn togamy: The two haploid pronuclei fuse, giving rise to an 1974), representing a haploid genome which has been entirely homozygous zygotic nucleus. In both cases, the variously estimated to contain between 45,000 and zygotic nucleus then divides twice; two of the products 280,000 kb (Gibson and Martin 1971; Cummings 1975; remain micronuclei while the two others differentiate McTavish and Sommerville 1980), should be much into two new macronuclei, which will segregate without longer on average. Because there appears to be little or no

GENES & DEVELOPMENT 6:211-222 © 1992 by Cold Spring Harbor Laboratory Press ISSN 0890-9369/92 $3.00 211 Downloaded from genesdev.cshlp.org on September 27, 2021 - Published by Cold Spring Harbor Laboratory Press

Meyer reduction in genome size during macronuclear develop­ alternative processing of the DNA during the develop­ ment (Allen and Gibson 1972; McTavish and Sommer- mental reorganization of the genome. ville 1980), chromosome breakage is very likely to occur, In this paper I show that an experimental modification leading to the formation of new telomeres in the macro- of the genetic content of the macronucleus by microin­ nucleus. jection of cloned DNA results in an altered processing of Different classes of reproducible heterogeneity in the specific micronuclear chromosomal regions during the position of the telomeres added during macronuclear dif­ following genome reorganization. When DNA molecules ferentiation have been observed. The first class is exem­ from various origins are microinjected into the macro- plified by the macronuclear chromosomes harboring the nucleus of P. aurelia, they are processed to yield stable G surface antigen gene. Formal genetic studies have linear chromosomes bearing telomeric repeats (C4A2 and shown that this gene is present in a single copy in the C3A3) at their ends that can replicate autonomously dur­ micronuclear genome (Beale 1952), yet it is found on at ing vegetative growth (Gilley et al. 1988), apparently least two different chromosomes in the macronucleus. without the need for a specific replication origin se­ One is 480 kb long, and the other is 250 kb long, with its quence. The injected DNA is lost at autogamy when the restriction map identical with that of the last 250 kb of macronucleus is degraded and replaced by a new one the 480-kb chromosome (Caron and Meyer 1989). The G derived from the micronucleus. When a plasmid contain­ gene lies near the telomere that is common to both chro­ ing the gene coding for surface antigen G of strain 156 mosomes. A second class of heterogeneity can be de­ (156G gene) of P. primaurelia is injected into the macro- fined, which involves smaller differences: Some chromo­ nucleus of strain 168 cells, the 156G protein can be sta­ somes show multiple telomeres, 5-15 kb apart, at one of bly expressed at the cell surface, as detected by an im­ their ends. Such is the case of the chromosome bearing mobilization test using specific antibodies (Caron and the A surface antigen gene of P. tetraurelia (Forney and Meyer 1989). After the following autogamy, the plasmid- Blackburn 1988). In contrast, the telomere located —5 kb encoded protein rapidly disappears. Since autogamy usu­ downstream of the G gene of P. phmaurelia is unique. ally does not affect serotype expression (autogamy of nat­ The other ends of both the 250- and the 480-kb chromo­ ural 156/168 heterozygotes expressing the G serotype, somes, however, show the multiple telomere structure e.g., yields F2 clones expressing either the 156 or the 168 (F. Caron, unpubl.). Finally, a third class of heterogeneity allele of the G gene), the transformed cell lines should is found within each of the telomeres, where the telo- maintain the expression of the endogenous 168G gene. meric repeats are added at different nucleotides within a However, it was unexpectedly observed that the endog­ region spanning 200-800 bp (Baroin et al. 1987; Forney enous 168G protein, like the plasmid-encoded 156G pro­ and Blackburn 1988). All three types of heterogeneity are tein, also disappears from the cell surface after autogamy found within caryonidal clones, that is, vegetative clones of the transformed cell lines. This observation prompted arising from a single macronuclear differentiation event. an investigation of the structure of the new macronu­ No conserved sequence element has been identified in clear genome, which showed that the disappearance of the vicinity of telomeric repeat addition sites in any of the 168G protein is due to the loss of the endogenous the telomere addition regions studied. Furthermore, a 168G gene from the new macronucleus. This work mutant has been described in P. tetraurelia in which shows that transformation of wild-type cells with a high telomeres are added to sequences within the gene coding copy number of the G-gene plasmid results in an alter­ for the A surface antigen, rather than at their normal native telomerization pattern of the homologous micro­ positions 8-26 kb from the 3' end of the gene. This re­ nuclear gene during the following macronuclear devel­ sults in the deletion from the macronucleus of the 3' part opment. of the gene and downstream sequences (Forney and Blackbum 1987). Thus, there appears to be no specific telomere addition sequence. What then determines the position of telomere addition regions during macronu­ Results clear differentiation? Induction of macronuclear mutations in strain 168 There is increasing evidence that the old macronu­ To see whether any change in the new macronuclear cleus itself is playing an important role in the control of genome might be induced by the transformation, DNA the genome reorganization. Some hereditary traits, such was extracted from transformed cell lines before and af­ as the mating type in P. tetraurelia (Sonneborn 1977) or ter autogamy. Plasmid pXA4 contains the entire tran­ a trichocyst discharge mutation (Sonneborn and scribed sequence of the 156G gene, as well as 230 bp of Schneller 1979), show cytoplasmic (non-Mendelian) in­ upstream sequences and 510 bp of downstream se­ heritance in genetic crosses, due to the determination of quences (see Fig. 1). The supercoiled form of this plasmid the new macronucleus by cytoplasmic factors produced was injected into the macronucleus of 168 cells; DNA by the old macronucleus. The d48 mutation also shows was extracted from the transformed cell lines and auto- cytoplasmic inheritance (Epstein and Forney 1984). The radiographs of Southern blots were scanned to estimate hereditary determinant of the mutation is located in the the amount of established plasmid. The copy number macronucleus, while the micronucleus is wild type was approximately 4,000, 24,000, 21,600, and 7,200 per (Harumoto 1986). In this case, it was shown that the old macronucleus for clones 1, 2, 3, and 4, respectively (Fig. macronucleus determines the new one by specifying an 2A,B), that is, between 5 and 30 times the normal

212 GENES & DEVELOPMENT Downloaded from genesdev.cshlp.org on September 27, 2021 - Published by Cold Spring Harbor Laboratory Press

Induction of developmental mutations in Paramecium

(a) B K

LK1A 156G

10 kb

(b) B K 156G

i > pEBGl 5' pPA2

1 kb pXA4 (c) B

(d) B

Figure 1. Map of the end of the wild-type G gene bearing macronuclear chromo­ (e) B 168G somes of strains 156 {a,b) and 168 (e,/). (c) Map of the first-generation mutant telo­ ~r- > mere (after autogamy of the injected cell DEfeGl-5' pP11 lines, both 156 and 168). {d) Map of the s5 mutant telomere. Arrows represent the transcribed sequences of the genes. Hatched boxes represent the telomere ad­ (f) dition regions. Black boxes represent the fragments used as probes for the Southern -( >—(d blots. The dotted line in b indicates the se­ LK1A 168G quence contained in plasmid pXA4.

amount of the gene. Seven clones that were injected with shows this defect, and neither does the post-autogamous a 1.6-kb linear £coRI fragment derived from the up­ clone derived from clone 1, which had a lower copy num­ stream part of the 156G gene were included as controls ber of the plasmid. Hybridization of the same blot or (lanes 5-11). The 1.6-kb fragment did not replicate au­ Southern blots of £coRI-digested DNA with various tonomously in the seven clones analyzed, as is shown by probes indicated that the entire coding sequence as well the absence of any hybridizing material at the appropri­ as the downstream sequences were missing from the mac- ate position in Figure 2B. However, this fragment was ronucleus of the mutant postautogamous clones (not integrated into the genomic DNA at a high copy number shown). Thus, transformation of 168 cells with a high in clone 9 through nonhomologous recombination (data copy number of plasmid pXA4 prevents the 168G micro- not shown; see Katinka and Bourgain 1992 for nonho­ nuclear gene from being correctly processed into the new mologous recombination of injected DNA), and possibly macronucleus after the following autogamy. in some other clones at a lower copy number. All 11 injected clones were allowed to undergo autogamy, and one or two independent postautogamous caryonides Characterization of the deletions were isolated for each clone and studied. Figure 2C The postautogamous clones lacking the 168G gene were shows a Southern blot of BamHI-digested DNA from further analyzed to determine the extent of the deletion. both pre- and postautogamous cell lines. The presence of A 1.9-kb Kpnl-EcoRl fragment located 71 kb upstream the 168G gene is revealed by hybridization with a 168- (orientation is defined relatively to the orientation of the specific probe, pPll (see Fig. 1). In wild-type cells, G gene; see Fig. 1) from the telomere of the G gene- BamHl generates a 13.5-kb telomeric fragment contain­ bearing macronuclear chromosome was cloned from a ing the 168G gene. This fragment is absent or much re­ jumping genomic library (F. Caron, in prep.). This frag­ duced in intensity in postautogamous clones derived ment, LKIA^ was used to probe a pulse-field electropho­ from clones 2, 3, and 4. (Micronuclear DNA only repre­ resis Southern blot of ^p22l-digested DNA (Fig. 3A). The sents 1/400 of the total cellular DNA and is not detect­ 72-kb telomeric fragment generated by Kpnl digestion in able on the Southern blots.) None of the postautogamous wild-type 168 cells and in the transformed cell lines is clones derived from clones injected with the 1.6-kb frag­ reduced in size to yield three main bands of about 60, 52, ment (or from uninjected control cells, not shown) and 43 kb. The relative intensities of these three bands

GENES & DEVELOPMENT 213 Downloaded from genesdev.cshlp.org on September 27, 2021 - Published by Cold Spring Harbor Laboratory Press

Meyer

A. 234056789 10 11 L-'D l1 234056789 10 11

kb .«»H «i

-"-•ft -10-

-5 -

-2 -

r1.6

- 1 -

i£ 1 1a 2 2a 3 3a 4 4a 4b 5 5a 6 6a 7 7a 8 8a 9 9a 10 ° ° 11 ;: ;:

kb I i -10

- 5 Figure 2. {A,B) Southern blot of a 0.6% agarose gel electrophoresis of uncut total DNA extracted from 168 clones injected with plasmid pXA4 (lanes l-4\ or with a 1.6-kb fragment from the upstream part of the 156G gene (lanes 5-21), probed with plasmid pXM. Lane 0 contains uninjected 168 DNA mixed with 0.5 ng of the 1.6-kb fragment as a control. Exposure was 2.7 hr [A] or 24 hr [B]. The arrows on the left indicate the position of the monomeric (M), dimeric (D), and trimeric (T) forms of the injected plasmid. The signal seen with genomic DNA in clone 9 is more intense than the background cross-hybridization of plasmid pXA4 with the genomic 168G gene sequences (lane 0] because of integration of multiple copies of the 1.6-kb fragment. (C) Southern blot of a 0.6% agarose gel electrophoresis of BamHI-digested DNA from the same injected cell lines and the postautogamous clones derived from them, probed with the 168-specific probe pPl 1. Numbering of the clones is as in A and B; the postautogamous clones are indicated by a or b. (Lane 168] Wild-type strain 168 DNA. The main band is the BamHl telomeric fragment containing the 168G gene (see Fig. 1). Faint bands visible above this band in clones 1, 2, 3, and 4 are due to hybridization of the injected plasmid with pUC18 sequences contaminating the purified probe pPl 1.

vary among different postautogamous clones; the largest between I and 2 kb, together with a faint and variable one, however, is always the most intense, except in band at 13.5 kb, corresponding to a very small amount of clone la, where it is absent. Other minor bands are also wild-type telomeres (also seen in Fig. 2C). Thus, the 60- seen below 43 kb in some clones. The simplest explana­ kb band results from the addition of telomeres to the tion of these results is that the G gene-bearing chromo­ region located immediately upstream of the insert of somes suffer terminal deletions leading to the formation plasmid pXA4. of new telomeres upstream from the wild-type telomere. If the 60-kb band truly represents a telomeric fragment, it should show the characteristic size heterogeneity (het­ Induction of macronuclear deletions in strain 156 erogeneity in the exact point of addition of the telomeric The transformation experiment was repeated with 156 repeats and in the number of repeats). The telomere cor­ recipient cells to see whether the overall sequence sim­ responding to the 60-kb band should be located a short ilarity between the injected plasmid and the endogenous distance downstream of the BamHI site, which is located gene had any effect on the phenomenon (the 156 and 168 58.5 kb downstream of the Kpnl site. Figure 3C shows a sequences corresponding to the insert of pXA4 are only Southern blot of BamHI-digested DNA from the mutant about 94% similar; Prat 1990). Injection of plasmid pXA4 postautogamous clones probed with a 1.9-kb fragment into the macronucleus of 156 cells also resulted in the immediately downstream from the BamHI site (probe production of postautogamous mutants unable to ex­ pEBGl; see Fig. I). The expected smear is indeed seen press the G gene. A molecular analysis of three of these

214 GENES & DEVELOPMENT Downloaded from genesdev.cshlp.org on September 27, 2021 - Published by Cold Spring Harbor Laboratory Press

Induction of developmental mutations in Parameciiun

A B

^ c 1 1a 2 2a 4 4a Jin2 i2n f g kb

— 84 _ WTi WT

— 42 _

— 12

Figure 3. {A,B] Southern blots of a pulse-field gel

OO (O electrophoresis (CHEF) of KpnI-digested DNA ^ a b c d 2a 3a 4a 4b i2 e f from pre- and postautogamous 168 clones {A] and postautogamous 156 clones [B], probed with frag­ ment LKIA. (C) Southern blot of a 0.6% agarose kb WT»- H"" ^^ •• H^ *^ If gel electrophoresis of BamHI-digested DNA from 10 168 and 156 postautogamous mutant cell lines, probed with fragment pEBGl. Numbering of the clones is as in Fig. 2. (Lanes a,b,c,d] Additional 168 mutants obtained after autogamy of a 168 clone injected with plasmid pXA4 in a previous experiment. (Lanes e,f,g) 156 mutants obtained af­ ter autogamy of a 156 clone injected with pXA4. (Lane 168) Wild-type 168 DNA; (lane 156] wild- type 156 DNA. Arrowheads point to the position M of wild-type (WT) or mutant main (M) telomeric fragments. The signal seen below 12 kb in lanes 1, 2, and 4 is due to hybridization of a Kpnl fragment W — 1 of the injected plasmid with the pUC18 moiety of the unpurified probe. The dotted line on the left of C indicates the smear due to the mutant telo-

mutants is presented in Figure 3, B and C. Here again, tions of internal sequences (between the Kpnl site and probing a Southern blot of a pulse-field gel of Kpnl-di- the mutant telomere). Thus, in spite of the complex pat­ gested DNA with fragment LKIA showed the same set of tern seen with pulse-field electrophoresis, there is only bands as was seen with the 168 mutants (Fig. 3B). Note one telomere addition region in the mutants. The addi­ that the band observed in wild-type 156 DNA is not a tional internal deletions will not be discussed any fur­ telomeric fragment, because of the presence of a second ther because (1) their occurrence was unpredictable Kpnl site in the coding sequence of the 156G gene (see (sometimes they were not seen at all) and (2) they were Fig. 1). The reduction in size from this 64-kb restriction no longer seen after a subsequent autogamy of the mu­ fragment to the largest (60-kb) telomeric fragment of the tant cell lines, whereas terminal deletions were still ob­ mutants is less obvious than in the case of the 168 mu­ servable (see section below. Stability of the macronu- tants. To determine the origin of the minor bands of 52 clear deletions). Figure 3C shows that the mutant telo­ and 43 kb, a blot identical to the one shown in Figure 33 mere addition region spans exactly the same sequence in was probed with fragment pEBGl; the same pattern was the 156 mutants as in the 168 mutants, between 1 and 2 again observed (not shown). Since the 52- and 43-kb frag­ kb downstream of the BarnHl site. This implies a 14-kb ments hybridize equally well with both probes LKIA and deletion instead of the 12-kb deletion seen in 168 mu­ pEBGl, which represent the two extremities of the 60-kb tants, the 156 wild-type telomere being located 2 kb far­ mutant telomeric fragment, they cannot be due to larger ther downstream than the 168 wild-type telomere (see terminal deletions, but rather represent additional dele­ Fig. 1).

GENES & DEVELOPMENT 215 Downloaded from genesdev.cshlp.org on September 27, 2021 - Published by Cold Spring Harbor Laboratory Press

Meyer

Specificity of the deletions The only phenotype displayed by the mutants is their A B inability to express the G serotype. Growth rate and morphology were indistinguishable from wild type. This 12 13 14 2 12 12a 14 14a argues against the idea of a generalized deletion of 12 kb or more from all macronuclear telomeres^ which would kb probably affect essential genes, given the relatively small !• —WT size of macronuclear telomeres. It is impossible to prove 10- that no other telomere is being affected; however, three M-- SIS -10 other macronuclear telomeres, for which probes were 5- available, were analyzed by pulse-field electrophoresis and shown to have the same pattern in wild-type and mutant cells (data not shown). Two of these (including -5 the other end of the 250 kb chromosome bearing the G Figure 4. [A] Southern blot of a 0.6% agarose gel electropho­ gene) show multiple telomere addition regions 4—20 kb resis of uncut total DNA extracted from 156 clones transformed apart. The third one, like the G gene telomere, has a with yeast plasmid YEpJBl-23-0 (lanes 13,14], and from one 168 single telomere addition region. Thus, it appears that the clone transformed with plasmid pXA4 (lane 2; this is clone 2 of deletion is specific for the G gene telomere. Fig. 2), which did yield deletions of the endogenous G gene after autogamy. Lane 12 is an uninjected control. The blot was The deletion is also a specific effect of the G gene probed with a 481-bp Xmnl-Bgll fragment of the ampicillin- plasmid. This is shown by a control experiment in which resistance gene, which is common to pXA4 and YEpJB 1-23-0, so a different plasmid, YEpJBl-23-0 (Banroques et al. 1987), that the copy numbers of the two plasmids can be compared was injected into 156 cells. This plasmid is a yeast shut­ directly. Scanning of an appropriate exposure indicated that the tle vector containing 2|JL and yeast genomic sequences copy number of clone 13 is about the same as that of clone 2, cloned into a pBR derivative; its size is similar to that of while that of clone 14 is twice as high. The arrowheads [left] plasmid pXA4 (11.5 kb). When the supercoiled form of indicate the position of the monomeric (M), dimeric (D), and this plasmid is injected into the macronucleus, the same trimeric (T) forms of the injected plasmids. (B) Southern blot of pattern of replicating monomers and multimers is ob­ a 0.8% agarose gel electrophoresis of BamHI-digested DNA ex­ served as following pXA4 injection (see Fig. 4A). Several tracted from clones 12 and 14 and the postautogamous clones cell lines transformed with this plasmid, with copy num­ derived from them (lanes 12a and 14a, respectively). The blot was probed with the 156G-specific probe pPA2. The arrowhead bers similar or twice as high as the highest copy number indicates the wild-type BamHl telomeric fragment containing of pXA4 in the first experiment (see Fig. 4A), were taken the G gene (WT). through autogamy. The processing of the micronuclear G gene was not affected: All postautogamous clones re­ tained the ability to express the 156G antigen, and telo­ meres were added at the wild-type position in all the clones analyzed (an example is shown in Fig. 4B). One of these clones (e5) was then taken through suc­ cessive autogamies. e5 had a grossly bimodal telomeric distribution, with two telomeric regions around 2 and 5 Stability of the macronuclear deletions kb downstream of the BamHl site. Figure 6 shows that a Two of the 156 mutants (e and f) were taken through a single clone isolated after a third autogamy (s) retained second autogamy to see whether the altered telomere only the telomere addition region around 5 kb. Clone s position could have any influence on the processing of was taken through a fourth autogamy: five independent the G gene during the following macronuclear develop­ postautogamous clones were analyzed (sl-s5), as well as ment. Figure 5 shows a Southern blot of BflmHI-digested a mass culture grown from 100 cells that were 97% au­ DNA probed with fragment pEBGl, which compares five togamous (sm). It is obvious from Figure 6 that the out­ independent postautogamous clones derived from mu­ come of this last autogamy is much less variable than tant e and 6 such clones from mutant f. The outcome of that of the autogamy of the mutants shown in Figure 5. autogamy in these mutants was highly variable, but gen­ All five independent postautogamous clones, as well as erally yielded a mix of mutant telomeres (resembling the mass autogamy of 100 cells, presented a similar telo­ those of mutants e and f) and wild-type telomeres, with meric distribution, slightly farther downstream than occasional multiple intermediate telomere positions. that of clone s. No wild-type telomeres were observed. One clone (f4) retained the same telomere distribution as Only s2 had a telomeric addition region significantly mutant f (autogamy was checked by staining three cells larger, extending 2-3 kb farther downstream. This vari­ out of four after the first division of the caryonides to ability is not significantly greater than the variability verify the presence of macronuclear fragments). Others observed after the autogamy of strain 168, where the (f2, f3, f6, e2) had a sizable amount of wild-type telo­ broadness of the telomeric distribution of different cary­ meres and were actually expressing G at the time of onides varies from 0.8 to 2 kb, while the average position DNA extraction. Some (el, e3) only had telomeres at of telomeric repeats varies by more than 1 kb (A.-M. intermediate positions. None of the nine clones had a Keller, A. Le Mouel, F. Caron, M.D. Katinka, and E. wild-type amount of wild-type telomeres. Meyer, in prep.).

216 GENES & DEVELOPMENT Downloaded from genesdev.cshlp.org on September 27, 2021 - Published by Cold Spring Harbor Laboratory Press

Induction of developmental mutations in Paramecium

I? f f1 f2 f3 f4 f5 f6 e el e2 e3 e4 e5

kb WTi -10

- 5

- 2 Figure 5. Southern blot of a 0.6% gel electropho­ resis of 5flmHI-digested DNA from mutants e and - 1 f and the postautogamous lines derived from them (lanes fl-f6,el-e5], probed with fragment pEBGl. (Lane 156] Wild-type 156 DNA. The arrowhead indicates the wild-type telomeric fragment (WT).

Macronuclear DNAs from clones of the mutant cell amy) nor in any of the following generations (data not lines after each autogamy (clones e5, s, s5, and the mass shown). Thus, the occasional internal deletions seen af­ autogamy sm) were also digested with Kpnl and analyzed ter autogamy of the injected cell lines, in contrast to the on a Southern blot of a pulse-field gel probed with frag­ terminal deletions, did not reproduce in the new macro- ment LKIA to check for the presence of the minor 52- nucleus after the following autogamy. and 43-kb bands that were seen in the first-generation mutant e, as in mutants f and g (see Fig. 3B). Those bands were no longer seen in clone e5 (after the second autog- Genetics of the macronuclear deletions

To ensure that the micronuclear genome had not been modified by the microinjection or by the successive au­ ^2 e e5 s s1 s2 s3 s4 s5 sm togamies of the mutants, the ability of the micronuclear kb G gene to be processed correctly was tested by crossing the 156 mutant cell lines obtained after the fourth auto­ -10 gamy with wild-type 168 cells. If the micronucleus of the mutants is wild type, the 156G gene should be normally processed in the heterozygote progeny of the 168 parent. lvf#f - 5 Mutant lines s2 and s5 were chosen because of the slight t difference in their telomeric distribution. They were f crossed with 168 cells expressing the G serotype (see Fig. 7 for a schematic representation of these crosses). The expression of both alleles was tested in each of the four caryonides after three or four divisions. Eight pairs were studied and showed expression of both alleles in the two caryonides derived from the 168 parent, while the two caryonides derived from the mutant expressed neither r the 156 nor the 168 allele. Figure 8 shows a molecular analysis of two such crosses. DNA was extracted from the four caryonides, as well as from the parents just be­ Figure 6. Southern blot of a 0.6% agarose gel electrophoresis of fore the cross, digested with BamHl, and hybridized on a BamHI-digested DNA from cell lines obtained through succes­ Southern blot successively with the 156-specific probe sive autogamies of mutant e, probed with fragment pEBGl. pPA2 (Fig. 8A) and the 168-specific probe pPll (Fig. 8B). (Lane e5] Postautogamous clone derived from e (same as in Fig. As expected from the expression pattern, both alleles 5); (lane s) postautogamous clone derived from e5; (lanes have telomeres at the wild-type position in the cary­ sl,s2,s3,s4,s5] five individual postautogamous clones derived onides derived from the 168 parent of both crosses. Thus, from S; (lane sm] culture grown from a mass autogamy of 100 the micronucleus of the mutants contains copies of the cells of clone S; (lane 156] wild-type 156 DNA. The arrowhead indicates the wild-type telomeric fragment (WT). G gene that can be correctly processed in a wild-type

GENES & DEVELOPMENT 217 Downloaded from genesdev.cshlp.org on September 27, 2021 - Published by Cold Spring Harbor Laboratory Press

Meyer

s5 mutant x WT 168 wild-type positions. This is true for both the 156 and the 168 alleles. Thus, the wild-type micronuclear 168G gene is not correctly processed in the mutant cell. Therefore, 1 /^O it can be concluded that these mutations have a macro- nuclear pattern of inheritance. Two additional observa­ tions can be made from this experiment. First, the dif­ ference in telomeric distribution between mutants s2 and s5 is reflected in the difference between the het- 156G- 168G+ erozygote macronuclei derived from each of them: For both alleles, the telomeric distribution of the heterozy- gote progeny of mutant s2 is farther downstream than that of mutant s5, just as the telomeric distribution of mutant s2 is farther downstream than that of mutant s5. Second, in a given heterozygote macronucleus the telo­ meric distribution is not the same for both alleles. It is always, on average, farther downstream for the 168 allele than for the 156 allele. On the blot shown in Figure 8A, the telomeric fragments of the 156 allele are less intense in the mutant s5 and the heterozygotes derived from this mutant than in the heterozygotes derived from the 168 parent. This is because the 156-specific probe, pPA2, cor­ responds to a sequence which is tandemly repeated five times in the coding sequence of the G gene. The telo­ mere addition region of mutant s5 coincides with this repeated structure, so that only the longest of the telo­ meric fragments of this mutant can be labeled as in­ tensely as the wild-type telomeric fragments. To quan­ tify the number of copies of each allele in the heterozy­ gotes, DNA was digested with BamHl and Hphl, which 56G+ gives a BamHl-Hphl fragment (downstream of the 1680+ BamHl site, but 5 kb upstream of the telomere addition region of mutant s5) of 880 bp for the 156 allele and 692 Figure 7. Schematic representation of the cross of the 156 mu­ bp for the 168 allele. The Southern blot shown in Figure tant s5 with wild-type 168. [1] The large ovals represent the 8C was probed with the corresponding part of fragment macronuclei and the circles represent the micronuclei. White pEBGl and shows that the number of copies of each al­ and black symbolize the 156 and 168 genomes, respectively. The phenotypes indicated beneath the cells refer to the ability lele in the four caryonides of both crosses is roughly half of the cells to express the two alleles of the G gene. (2) After that of the corresponding allele in the parents. Note that pairing of the cells, the two micronuclei of each cell enter mei- the probe used was derived from the 156 allele and does osis. Of the eight haploid products, seven degenerate and the not hybridize as well with the 692-bp fragment of the remaining one divides once more to yield two identical haploid 168 allele. pronuclei, which are represented by small circles. The two mat­ ing cells then exchange one of their pronuclei. During this time the old macronucleus is fragmented. The fragments are repre­ sented clumped together for the sake of clarity. (3) Fusion of the Discussion exchanged pronucleus with the remaining one in each cell re­ sults in the formation of an identical heterozygous nucleus in In this paper it was shown that transformation of P. ph- the two mates. Heterozygous nuclei are depicted in gray. [4] The maurelia with a high copy number of a plasmid contain­ zygote nucleus then divides twice. Two of the products remain ing the G gene specifically modifies the position at micronuclei, while the other two begin to differentiate into mac­ which telomeres are formed during the following macro- ronuclei (small ovals). (5) At the first division after conjugation, nuclear differentiation. The alternative telomere posi­ the two new macronuclei of each exconjugant segregate with­ out division to the two daughter cells, called caryonides, while tion results in the almost complete elimination of the G the two micronuclei divide by mitosis. The phenotype of the gene from the new macronucleus, producing mutant four caryonides obtained from this cross (al, a2, a3, a4), corre­ progeny unable to express the G serotype. It is striking sponding to four independent differentiation events of identical that the mutant telomere position is immediately up­ micronuclei, illustrates the influence of the fragments of the old stream of the sequence contained in plasmid pXA4 in macronucleus on the differentiation process. both 168 and 156 mutants, although this implies dele­ tions of different lengths from the two wild-type telo­ meres. Whether this is a coincidence or a consequence of cell. In the caryonides derived from the mutant, the the mechanisms involved remains to be determined. Ex­ telomeres occupy multiple positions that are similar to periments are in progress to determine whether injection the mutant or intermediate between the mutant and the of plasmids carrying different inserts can induce dele-

218 GENES & DEVELOPMENT Downloaded from genesdev.cshlp.org on September 27, 2021 - Published by Cold Spring Harbor Laboratory Press

Induction of developmental mutations in Paramecium

S5x 168 s2x168 •>- •<- ^ s5 a1 a2 a3 a4 ^ s2 b1 b2 b3 b4 ^ Figure 8. Molecular analysis of the crosses of 156 mu­ tants s5 and s2 with the wild-type 168 strain. [A] kb Southern blot of a 0.6% agarose gel electrophoresis of WT-^I -10 BamHI-digested DNA from the two parent clones and WKF ^^ the four caryonides of each cross, probed with the 156- A specific probe pPA2. (Lanes al,a2] Caryonides derived from the mutant s5 parent; (lanes a3,a4) caryonides derived from the wild-type 168 parent; (lanes bl,b2) caryonides derived from the mutant s2 parent; (lanes b3,b4] caryonides derived from the wild-type 168 par­ WT^ ent; (lane 156] wild-type 156 DNA (this lane inadvert­ -10 ently contains twice as much DNA as the others). The w arrowhead points to the wild-type telomeric fragment B (WT). [B] Same blot as in A, dehybridized and rehybrid- ized with the 168-specific probe pPll. (C) Southern -5 blot of a 1 % agarose gel electrophoresis of BamHl- and Hphl-digested DNA from the same clones, probed with a 467-bp fragment located 208-675 bp downstream of the BamHl site of the 156 allele. The 156 lane contains twice as much DNA as the others. Arrowheads point to the position of the 880- and 692-bp BamHl-Hphl fragments, respectively, derived from the 156 and the 168 alleles.

tions at other points of the sequence. The mutant telo­ same micronuclear genome, yet telomeres rcproducibly mere position is not stable through a second autogamy: form at different positions of the G gene sequence at Telomeres nov\^ form at a variety of different positions in each autogamy. Whether or not the G gene macronuclear different caryonides. However, the mutant telomere still telomere is produced by a breakage of the micronuclear exerts a strong influence on the processing of the micro- chromosome during macronuclear development, the dif­ nuclear copy of the G gene, because few telomeres, if ferent telomere positions in the two strains cannot be any, are formed at the wild-type position in each cary- determined by a cis-acting element such as the chromo­ onide. The telomere position can be stabilized after a some breakage sequence (Cbs) recently identified in Tet- third autogamy, as shown by the mass autogamy exper­ rahymena (Yao et al. 1990). Clearly, in this case, another iment, in which the pattern obtained from a mass culti­ mechanism is involved in the positioning of the telo­ vation of 100 independent autogamous cells is similar to meres, which is determined by the old macronucleus that of the parent. Crosses of the stabilized mutants with rather than by the micronuclear sequence. Therefore, a wild-type cells show that their micronuclei contain cop­ specific signal has to be transmitted from the old macro- ies of the G gene that can still be normally processed nucleus to the developing one. This conclusion had al­ when the new macronucleus develops in a wild-type ready been reached in the study of the d48 mutation after cell. Conversely, the micronuclear copies of wild-type it was shown that transplantation of macronucleoplasm cells are not normally processed in the mutant cells. from wild-type cells into the macronucleus of d48 al­ Therefore, those mutations show the same macronu- lowed the mutant cells to revert permanently to wild clear pattern of inheritance as d48, a previously de­ type after the following autogamy (Harumoto 1986). It scribed deletion mutant of P. tetraurelia (Epstein and was further shown that microinjection of a plasmid con­ Forney 1984), and confirm that this pattern of inheri­ taining the A gene into the macronucleus had the same tance results from the influence of the old macronucleus effect; transfer of cytoplasm from wild-type cells to d48 on the process of developmental reorganization of the cells also rescued the d48 mutation if donor and recipi­ genome. ent cells were at specific stages in autogamy (Koizumi and Kobayashi 1989). Rescuing plasmids do not neces­ sarily have to contain the whole A gene sequence: A Influence of the macronucleus on the processing plasmid lacking the last 1 kb of the coding sequence was of the macronuclear genome able to rescue d48; however a plasmid containing only 2.1 kb from the middle of the coding sequence did not Although the data presented in this report are not suffi­ have this ability (You et al. 1991). These experiments cient to propose a general model explaining how the pro­ have led to the hypothesis that a specific factor is pro­ cessing of micronuclear genes is affected by the content duced by a defined region of the macronuclear A gene of the old macronucleus, some important points can be and acts via the cytoplasm on the developing macronu­ made. The wild-type 156 and s mutant strains have the cleus to ensure normal processing of the A gene.

GENES & DEVELOPMENT 219 Downloaded from genesdev.cshlp.org on September 27, 2021 - Published by Cold Spring Harbor Laboratory Press

Meyer

Evidence for mutant processing factors telomeric distribution). The heterogeneity in the postau- togamous telomeric distributions could be linked to that However, the question of what determines the position preautogamous condition, if we assume that only a ran­ of the telomeres in the mutant macronucleus has not dom subset of the processing factors produced by a ma­ been addressed. The simplest explanation for the stable cronucleus with a heterogeneous telomeric distribution differences in telomere position between wild-type and is used for the determination of the new macronucleus. mutant cell lines, both in P. tetraurelia and in P. primau- In addition, different processing factors might have dif­ relia, is that the telomeres are positioned in the devel­ ferent efficiencies that would result in different telomere oping macronucleus just as they are in the old macronu­ positions having different stabilities, leading to unpre­ cleus. I propose that each macronuclear telomeric region dictable competition. The stabilization of the mutant is able to produce a specific factor that directs the pro­ telomere position after several successive rounds of au­ cessing of the corresponding micronuclear sequence by togamy could thus simply be due to the random selec­ transmitting information about the position of telo­ tion of a particular telomere position. Under this hypoth­ meres to the developing macronucleus. Several lines of esis, the high frequency of spontaneous reversion of the evidence support this idea. (1) The fact that wild-type P. d48 mutation at autogamy (You et al. 1991) could be primaurelia cells transformed with pXA4 (but not unin­ linked to the reported presence of a small number of fected controls nor cells transformed with a different rep­ wild-type telomeres in the d48 macronucleus. It should licating plasmid) produce progeny carrying a specific ma­ also be noted that even when autogamy alters the pat­ cronuclear deletion of the G gene suggests that a specific tern of telomeric distribution, a faint image of the old factor produced from the telomerized plasmid is respon­ macronucleus distribution can still be seen in the new sible for this deletion. In contrast to the d48 rescue ex­ one (see Figs. 3C, 5, and 6), so that even unstable telo­ periments, the specific processing factor is needed here mere positions seem to be able to reproduce in the new to explain formation of the telomeres at a mutant posi­ macronucleus. However, the appearance of telomere po­ tion. This factor could override the effect of the factor sitions intermediate between the mutant and the wild- normally produced from the wild-type telomere, still type positions after the autogamy of the first-generation present in normal amounts in the macronucleus of in­ mutants remains difficult to explain. jected cells before autogamy, because of the high copy number of the plasmid. Indeed, in the experiment shown in Figure 2, the ratio of telomerized plasmid to wild-type Predetermination of the micronuclei telomeres before autogamy seems to be correlated with by the processing factorsf- the ratio of mutant to wild-type telomeres after autog­ amy. (2) In the course of this study, many different The crosses of the stabilized 156 mutants with strain 168 telomere positions have been observed along the G gene- show that the telomeric distributions of both alleles are coding and flanking sequences, which is consistent with heterogeneous in the heterozygote caryonides derived previous observations that there appears to be no specific from the mutant parent. This appears to be at odds with sequence requirement for the formation of telomeres. It the fact that autogamy of the 156 mutants introduces was not proven that all telomere positions can be stable; little or no heterogeneity in the telomeric distribution. however, the two crosses presented in Figure 8 show that How could the mutant macronucleus affect differently mutants s2 and s5, which differ only slightly in the po­ the processing of the 156 allele in the homozygous nu­ sition of the telomeres, influence the processing of the clei of autogamous mutant cells and in the heterozygous micronuclear 168G gene differently, the telomere posi­ nuclei of the mutant exconjugants? One possibility is tion in the mutant parent defining the upstream-most that the haploid pronuclei are at least partially predeter­ telomere position in the heterozygotes. Thus, these two mined by the processing factors even before they are ex­ mutants produce different specific processing factors. changed between the mating cells. This is consistent with the timing of production of the processing factors determined by the cytoplasm transfer experiments of Koizumi and Kobayashi (1989). They showed that the Stability of the macronuclear telomeric distribution ability of cytoplasm from wild-type autogamous cells to What, then, would determine the stability of any partic­ rescue the d48 mutation could already be detected at the ular macronuclear telomeric distribution through auto­ stage of lobed macronucleus, well before the stage cor­ gamy? It was shown that the outcome of the autogamy of responding to the exchange of the pronuclei in conjuga­ the first generation mutants (those obtained after the tion. Processing of the genes in the developing heterozy­ autogamy of the injected cell lines) was variable (Fig. 5). gote macronuclei of the mutant exconjugant would then Autogamy of the second generation mutant e5 also al­ introduce the same heterogeneity as is seen in the ma­ tered the telomeric distribution (Fig. 6). Although this cronuclei developing in an autogamous cell where the does not seem to support the idea that telomeres always old macronucleus shows a heterogeneous telomeric dis­ reproduce themselves at the same position, it should be tribution. The fact that the telomeric distribution of the noted that in both cases the telomere position was not 168 alleles in these heterozygote macronuclei is different unique in the macronucleus of the cells that underwent from that of the 156 alleles, and closer to the wild-type, autogamy (first-generation mutants still retain some may be related to the predetermination of micronuclear wild-type telomeres, while mutant e5 has a bimodal copies. Perhaps this could also explain the occasional

220 GENES & DEVELOPMENT Downloaded from genesdev.cshlp.org on September 27, 2021 - Published by Cold Spring Harbor Laboratory Press

Induction of developmental mutations in Paramecium formation of wild-type telomeres in the caryonides de­ tion test after two more divisions, and further confirmed after rived from a d48 cell crossed with a wild-type cell. The DNA extraction by the identification of the mitochondrial asymmetry of these crosses (macronuclear development DNA restriction pattern, which is strain specific. on the wild-type side does not seem to be influenced by the micronucleus coming from the mutant cell) could Microinjections mean that the processing factors corresponding to the wild-type telomere position have a dominant effect, or Caryonidal clones of 15-30 fissions of age were injected in are more efficient than others, giving a greater stability Dryl's medium (Dryl 1959) containing 0.2% BSA, under an oil to wild-type telomeres. Interestingly, an asymmetrical film, while being visualized with a phase-contrast inverted mi­ croscope (Axiovert 35M, Zeiss). Approximately 5 pi of a solu­ predetermination of the micronuclei had already been tion of CsCl-purified plasmid DNA, that had been filtered on a postulated to explain deviations from the normally ma­ 0.22-jjLm Millex-GV4 filter (Millipore) and disolved in water at a cronuclear inheritance of the mating type differentiation concentration of 5 mg/ml, was delivered into the macronucleus. in P. tetraurelia (Brygoo et al. 1980). If indeed each macronuclear telomeric region is able to produce a specific factor, and since there appears to be no DNA extraction stringent sequence requirement for the formation of ma­ Cultures (400 ml) of exponentially growing cells at 1000 cells/ cronuclear telomeres, it is difficult to escape the conclu­ ml were centrifuged. After being washed in Dryl's medium, the sion that the factors are the telomeric regions them­ pellet was resuspended in one volume of the same buffer, and selves, or a copy of them. Whereas the molecular nature quickly added to four volumes of lysis solution [0.44 M EDTA of the processing factors remains unknown, a careful (pH 9.0), 1% SDS, 0.5% N-laurylsarcosine (Sigma), and 1 mg/ml study of what is required for the injected plasmid to in­ proteinase K (Merck)] at 55°C. The lysate was incubated at 55°C for at least 5 hr, gently extracted once with phenol, and dialyzed duce deletions (copy number, form, sequence, patterns of twice against TE (10 mM Tris-HCl, 1 mM EDTA (pH 8.0)) con­ linearization, and telomerization) may help define their taining 20% ethanol, and once against TE. role in the processing of the micronuclear genes. What­ ever this role may be, it is now possible to induce exper­ imentally specific modifications of the telomerization DNA restriction and electrophoresis pattern during macronuclear differentiation. These procedures were carried out according to standard meth­ ods (Sambrook et al. 1989). Pulse-field electrophoresis was car­ ried out in a contour-clamped homogeneous electric field Materials and methods (CHEF) home-made apparatus (Chu et al. 1986) in 0.25 x THE Cell lines and cultivation (THE is 89 mM Tris, 89 mM borate, and 2.5 mM EDTA) at 5 V/cm, with cooling to 8°C and a commutation period of 4 sec. P. primaurelia wild-type strains 156 and 168 (Sonneborn 1974) are well-characterized stocks that have been used extensively in genetic studies of the surface antigen system. Cells were grown Southern hybridization in a grass infusion medium bacterized the day before use with Klebsiella pneumoniae and supplemented with 0.8 mg/liter of DNA was transferred from agarose gels to Hybond N^ mem­ p-sitosterol (Merck, Darmstadt, Germany). Cultivation, autog­ branes (Amersham, UK) in 0.4 N NaOH after depurination in amy, and conjugation were all carried out at 24°C. Basic meth­ 0.25 N HCl. Hybridization was carried out according to Church ods of cell culture have been described (Sonneborn 1970). and Gilbert (1984) in 7% SDS, 0.5 M sodium phosphate, 1% BSA, and 1 mM EDTA (pH 7.2) at 63°C. Probes were labeled using a random priming kit (Boebringer Mannheim) to a specific Injected DNA activity of 3 x 10^ cpm/^-g- Membranes were then washed for 30 min in 0.2 x SSC (SSC is 0.15 M NaCl and 0.015 M sodium Plasmid pXA4 is a derivative of pXI3, a larger plasmid contain­ citrate) and 0.5% SDS at 63°C (allele-specific probes pPA2 and ing the G gene from strain 156 (Caron and Meyer 1985), and pPll) or 60°C (other probes) prior to autoradiography. contains the entire transcribed region (8218 bp) as well as 230 bp of upstream and about 510 bp of downstream flanking se­ quences. The sequence of the whole insert was determined and partly published (Prat et al. 1986). Acknowledgments I thank Francois Caron for the gift of probe LKIA; Alain Butler, Francois Caron, Michael D. Katinka, and Linda D. Martin for Autogamy and conjugation critically reading the manuscript; Anne-Marie Keller and Anne Autogamy was induced by starving the cells after they had Le Mouel for sharing unpublished data; and all of the above for reached the appropriate clonal age, and assessed by staining stimulating discussions. Plasmid YEpJB 1-23-0 was kindly given with a 15 : 1 (vol/vol) mixture of carmine red (0.5% in 45% by Agnes Delahodde. This work was supported by grant 88/1470 acetic acid) and fast green (1% in ethanol). Cells were isolated from the Direction des Recherches, Etudes et Techniques, Min- from depressions showing >50% autogamous cells; autogamy istere de la Defense, and by a grant from the Ministere de la was checked by staining three cells out of four after the second Recherche et de I'Education Superieure (Structure et Fonction division of the isolated cell to verify the presence of macronu­ des Macromolcules et Systemes Integres). clear fragments. Conjugation was induced by starving two The publication costs of this article were defrayed in part by clones with complementary mating types. The four caryonides payment of page charges. This article must therefore be hereby from individual pairs were isolated; the cytoplasmic origin of marked "advertisement" in accordance with 18 USC section each of the caryonides was determined by an early immobiliza­ 1734 solely to indicate this fact.

GENES & DEVELOPMENT 221 Downloaded from genesdev.cshlp.org on September 27, 2021 - Published by Cold Spring Harbor Laboratory Press

Meyer

References Preer, J.R., Jr. and L.B. Preer. 1979. The size of macronuclear DNA and its relationship to models for maintaining genie Allen, S. and I. Gibson. 1972. Genome amplification and gene balance. /. Protozool. 26: 14-18. expression in the ciliate macronucleus. Biochem. Genet. Sambrook, J., E.F. Fritch, and T. Maniatis. 1989. Molecular clon­ 6:293-313. ing: A laboratory manual, 2nd ed. Cold Spring Laboratory Banroques, J., J. Perea, and C. Jacq. 1987. Efficient splicing of Press, Cold Spring Harbor, New York. two yeast mitochondrial introns controlled by a nuclear-en­ Sonneborn, T.M. 1970. Methods in Paramecium research. coded maturase. EMBO /. 6: 1085-1091. Methods Cell Physiol. 4: 241-339. Baroin, A., A. Prat, and F. Caron. 1987. Telomeric site position . 1974. Paramecium aurelia. In Handbook of genetics: heterogeneity in macronuclear DNA of Paramecium pri- Plants, plant viruses and protists (ed R.C. King), vol. 2, pp. maurelia. Nucleic Acids Res. 15: 1717-1728. 469-594. Plenum Press, New York. Beale, G.H. 1952. Antigen variation in Paramecium aurelia, . 1977. Genetics of cellular differentiation: Stable nuclear variety 1. Genetics 37: 62-74. differentiation in eucaryotic unicells. Annu. Rev. Genet. Brygoo, Y., T.M. Sonneborn, A.-M. Keller, R.V. Dippel, and M.V. 11:349-367. Schneller. 1980. Genetic analysis of mating type differenti­ Sonneborn, T.M. and M.V. Schneller. 1979. A genetic system for ation in Paramecium tetraurelia. II. Role of the micronuclei alternative stable characteristics in genomically identical in mating type differentiation. Genetics 94: 951-959. homozygous clones. Dev. Genet. 1: 21-46. Caron, F. and E. Meyer. 1985. Does Paramecium primaurelia Yao, M.-C. 1989. Site-specific chromosome breakage and DNA use a different genetic code in its macronucleus? Nature deletion in ciliates. In Mobile DNA. American Society for 314: 185-188. Microbiology, Washington, D.C. . 1989. Molecular basis of surface antigen variation in Yao, M.-C, C.-H. Yao, and B. Monks. 1990. The controlling paramecia. Annu. Rev. Microbiol. 43: 23-42. sequence for site-specific chromosome brekage in Tetrahy­ Chu, G., D. Vollrath, and R.W. Davis. 1986. Separation of large mena. Cell 63: 763-772. DNA molecules by contour-clamped homogenous electric You, Y., K. Aufderheide, J. Morand, K. Rodkey, and J. Forney. fields. Science 234: 1582-1585. 1991. Macronuclear transformation with specific DNA frag­ Church, G.M. and W. Gilbert. 1984. Genomic sequencing. Proc. ments controls the content of the new macronuclear ge­ Natl. Acad. Sci. 81: 1991-1995. nome in Paramecium tetraurelia. Mol. Cell. Biol. 11: 1133- Cummings, D.J. 1975. Studies on macronuclear DNA from Par­ 1137. amecium aurelia. Chromosoma 53: 191-208. Dryl, S. 1959. Antigenic transformation in Paramecium aurelia after homologous antiserum treatment during autogamy and conjugation. /. Protozoal. (Suppl.) 6: 25. Epstein, L.M. and J.D. Forney. 1984. Mendelian and non-Men- dclian mutations affecting surface antigen expression in Par­ amecium tetraurelia. Mol. Cell. Biol. 4: 1583-1590. Forney, f.D. and E.H. Blackburn. 1988. Developmcntally con­ trolled telomere addition in wild-type and mutant parame­ cia. Mol. Cell. Biol. 8: 251-258. Gibson, I. and N. Martin. 1971. DNA amounts in the nuclei of Paramecium aurelia and Tetrahymena pyriformis. Chromo­ soma 35: 374-382. Gilley, D., J.R. Preer, Jr., K.J. Aufderheide, and B. Polisky. 1988. Autonomous replication and addition of telomerelike se­ quences to DNA microinjected into Paramecium tetraurelia macronuclei. Mol. Cell. Biol. 8: 4765-4772. Harumoto, T. 1986. Induced change in a non-Mendclian deter­ minant by transplantation of macronucleoplasm in Parame­ cium tetraurelia. Mol. Cell. Biol. 6: 3498-3501. Katinka, M.D. and F.M. Bourgain. 1992. Interstitial telomeres are hotspots for illegitimate recombination with DNA mol­ ecules injected into the macronucleus of Paramecium pri­ maurelia. EMBO /. 11: (in press). Koizumi, S. and S. Kobayashi. 1989. Microinjection of plasmid DNA encoding the A surface antigen of Paramecium tetrau­ relia restores the ability to regenerate a wild-type macronu­ cleus. Mol. Cell. Biol. 9: 4398-4401. McTavish, C. and J. Sommerville. 1980. Macronuclear DNA organization and transcription in Paramecium primaurelia. Chromosoma 78: 147-164. Prat, A. 1990. Conserved sequences flank variable tandem re­ peats in two alleles of the G surface protein of Paramecium primaurelia. J. Mol. Biol. 211: 521-535. Prat, A., M. Katinka, F. Caron, and E. Meyer. 1986. Nucleotide sequence of the Paramecium primaurelia G surface protein. A huge protein with a highly periodic structure. /. Mol. Biol. 189: 47-60.

222 GENES & DEVELOPMENT Downloaded from genesdev.cshlp.org on September 27, 2021 - Published by Cold Spring Harbor Laboratory Press

Induction of specific macronuclear developmental mutations by microinjection of a cloned telomeric gene in Paramecium primaurelia.

E Meyer

Genes Dev. 1992, 6: Access the most recent version at doi:10.1101/gad.6.2.211

References This article cites 26 articles, 9 of which can be accessed free at: http://genesdev.cshlp.org/content/6/2/211.full.html#ref-list-1

License

Email Alerting Receive free email alerts when new articles cite this article - sign up in the box at the top Service right corner of the article or click here.

Copyright © Cold Spring Harbor Laboratory Press